In multiple antenna systems that use spatial multiplexing to raise transmission rates, it is preferable to use maximum likelihood (ML) detection to exploit the full receive diversity and minimize the error probability. In this paper, we present two tree based approximate ML detectors that use new two ordering criteria in conjunction with efficient search strategies. Unlike conventional tree detectors, the new detectors closely approximate the error performance of the exact ML detector while achieving a dramatic reduction in complexity. Moreover, they ensure a fixed detection delay and high level of parallelization in the tree search.

A modified Schnorr-Euchner sphere decoding (SE-SD) algorithm to search for the receive minimum distance is presented. In the proposed algorithm, the visit to negative symmetric vectors of already spanned vectors is avoided by using a biased spanning, and the redundant processes to visit the all-zero vector are also eliminated. A numerical experiment shows that the modified SE-SD algorithm is much more efficient than the conventional algorithm in terms of average computational complexity.

The mixture modeling framework is widely used in many applications. In this paper, we propose a component reduction technique, that collapses a Gaussian mixture model into a Gaussian mixture with fewer components. The EM (Expectation-Maximization) algorithm is usually used to fit a mixture model to data. Our algorithm is derived by extending mixture model learning using the EM-algorithm. In this extension, a difficulty arises from the fact that some crucial quantities cannot be evaluated analytically. We overcome this difficulty by introducing an effective approximation. The effectiveness of our algorithm is demonstrated by applying it to a simple synthetic component reduction task and a phoneme clustering problem.

Frequency offset estimation is an important technique in receiver design of wireless communications. In many applications, sampled single frequency tone is selected as training symbol/sequence for frequency synchronization. Under this assumption, frequency offset estimation can be regarded as the problem of single carrier frequency offset estimation. In this Letter, an approximate maximum likelihood frequency estimator is proposed. This estimator is efficient at moderate and high SNR's. Compared with other estimators, the proposed estimator is less sensitive to the variance threshold and offers feasible levels of computation complexity. The proposed estimator is suitable for high frequency offset cases and coarse/fine frequency synchronization applications.

The task of joint channel and data estimation based on the maximum likelihood principle is addressed using a continuous and discrete particle swarm optimization (PSO) algorithm over additive white Gaussian noise channels. The PSO algorithm works at two levels. At the upper level continuous PSO estimates the channel while at the lower level, discrete PSO detects the data. Simulation results indicate that under the same conditions, PSO outperforms the best of the published alternatives.

This paper considers the problem of target location estimation in a wireless sensor network based on IEEE 802.15.4 radio signals and proposes a novel implementation of the maximum likelihood (ML) location estimator based on the Cross-Entropy (CE) method. In the proposed CE method, the ML criterion is translated into a stochastic approximation problem which can be solved effectively. Simulations that compare the performance of a ML target estimation scheme employing the conventional Newton method and the conjugate gradient method are presented. The simulation results show that the proposed CE method provides higher location estimation accuracy throughout the sensor field.

In a centralized localization scenario, the limited throughput of the central node constrains the possible number of target node locations that can be estimated simultaneously. To overcome this limitation, we propose a method which effectively decreases the traffic load associated with target node localization, and therefore increases the possible number of target node locations that can estimated simultaneously in a localization system based on received signal strength indicator (RSSI) and maximum likelihood estimation. Our proposed method utilizes a threshold which limits the amount of forwarded RSSI data to the central node. As the threshold is crucial to the method, we further propose a method to theoretically determine its value. We experimentally verified the proposed method in various environments and the experimental results revealed that the method can reduce the load by 32-64% without significantly affecting the estimation accuracy.

Code acquisition performance in the Direct-Sequence Code-Division Multiple-Access (DS/CDMA) communication system is strongly related to the quality of the communication systems. The performance is assessed by (i) code acquisition time; (ii) precision; and (iii) complexity for implementation. This paper applies the method of maximum likelihood (ML) to estimation of propagation delay in DS/CDMA communications, and proposes a low-complexity method for code acquisition. First, a DS/CDMA system model and properties of outputs with a passive matched-filter receiver are reviewed, and a statistical problem in code acquisition is mentioned. Second, an error-controllable code acquisition method based on the maximum likelihood is discussed. Third, a low-complexity ML code acquisition method is proposed. It is shown that the code acquisition time with the low-complexity method is about 1.5 times longer than that with the original ML method, e.g. 13 data periods under 4.96 dB.

In this paper, we analyze the effects of IQ (In-phase/Quadrature-phase) imbalance at both transmitter and receiver of OFDM (Orthogonal Frequency Division Multiplexing) system and show that more diversity gain can be achieved even though there are unwanted IQ imbalance. When mixed sub-carriers within an OFDM symbol due to the IQ imbalance undergo frequency selective channels, additional diversity effects are expected during the demodulation process. Simulation results on the symbol error rate (SER) performance with ML (Maximum Likelihood) and OSIC (Ordered Successive Interference Cancellation) receiver show that significant performance gain can be achieved with the diversity gain caused by the IQ imbalance combined with the frequency selective channels.

Target location estimation is one of many promising applications of wireless sensor networks. However, until now only few studies have examined location estimation performances in real environments. In this paper, we analyze the effect of walking people on target location estimation performance in three experimental locations. The location estimation is based on received signal strength indicator (RSSI) and maximum likelihood (ML) estimation, and the experimental locations are a corridor of a shopping center, a foyer of a conference center and a laboratory room. The results show that walking people have a positive effect on the location estimation performance if the number of RSSI measurements used in the ML estimation is equal or greater than 3, 2 and 2 in the case of the experiments conducted in the corridor, foyer and laboratory room, respectively. The target location estimation accuracy ranged between 2.8 and 2.3 meters, 2.5 and 2.1 meters, and 1.5 and 1.4 meters in the case of the corridor, foyer and laboratory room, respectively.

This letter proposes a low-complexity estimation method of integer frequency offset in orthogonal frequency division multiplexing (OFDM) systems. The performance and complexity of the proposed method are compared with that of Morelli and Mengali's method based on maximum likelihood (ML) technique. The results show that the performance of the proposed method is comparable to that of M&M method with reduced complexity.

This paper investigates a scheme to improve a location estimation method for higher estimation accuracy in sensor networks. For the location estimation method, we focus on the maximum likelihood method based on the measurements of received signal strength and its known probability distribution. Using some statistical properties of the estimate obtained by the maximum likelihood method in a simplified situation, we propose a modification of likelihood function in order to improve the estimation accuracy for arbitrary situation. However, since the proposed scheme is derived under a special assumption for the simplification, we should examine the impact of the proposed scheme in more general situations by numerical simulation. From the simulation results, we show the effectiveness of the proposed modification especially in the cases of small number of samples (namely, the measurements of received signal strength) and the channel model with exponential distribution.

This paper describes a novel parameter generation algorithm for an HMM-based speech synthesis technique. The conventional algorithm generates a parameter trajectory of static features that maximizes the likelihood of a given HMM for the parameter sequence consisting of the static and dynamic features under an explicit constraint between those two features. The generated trajectory is often excessively smoothed due to the statistical processing. Using the over-smoothed speech parameters usually causes muffled sounds. In order to alleviate the over-smoothing effect, we propose a generation algorithm considering not only the HMM likelihood maximized in the conventional algorithm but also a likelihood for a global variance (GV) of the generated trajectory. The latter likelihood works as a penalty for the over-smoothing, i.e., a reduction of the GV of the generated trajectory. The result of a perceptual evaluation demonstrates that the proposed algorithm causes considerably large improvements in the naturalness of synthetic speech.

We develop a maximum likelihood estimation scheme for correcting the carrier frequency offsets prior to the general intercarrier interference (ICI) self-cancellation in the OFDM systems. Since the same data symbols employed for ICI self-cancellation are also used for frequency offset estimation, the proposed scheme does not consume additional bandwidth. The combined use of the estimation algorithm and ICI self-cancellation scheme provides both frequency offset compensation and ICI reduction hence improves the system performance greatly. The effectiveness of the proposed estimation-cancellation scheme is further verified by calculating the bit error rates of various OFDM receivers, and substantial improvements are found.

A new method to approximate the receive minimum distance is presented. In the proposed approximation, the geometric mean of the singular values of the channel matrix is used instead of the conventional minimum singular value. Numerical experiments show that the proposed approximation has less mean squared error than the minimum singular value bound and outperforms the minimum singular value bound in terms of bit error rate when they are applied to the antenna subgroup selection system.

We compare the convergence performance of different numerical schemes for computing the fundamental matrix from point correspondences over two images. First, we state the problem and the associated KCR lower bound. Then, we describe the algorithms of three well-known methods: FNS, HEIV, and renormalization. We also introduce Gauss-Newton iterations as a new method for fundamental matrix computation. For initial values, we test random choice, least squares, and Taubin's method. Experiments using simulated and real images reveal different characteristics of each method. Overall, FNS exhibits the best convergence properties.

In speaker adaptation for speech synthesis, it is desirable to convert both voice characteristics and prosodic features such as F0 and phone duration. For simultaneous adaptation of spectrum, F0 and phone duration within the HMM framework, we need to transform not only the state output distributions corresponding to spectrum and F0 but also the duration distributions corresponding to phone duration. However, it is not straightforward to adapt the state duration because the original HMM does not have explicit duration distributions. Therefore, we utilize the framework of the hidden semi-Markov model (HSMM), which is an HMM having explicit state duration distributions, and we apply an HSMM-based model adaptation algorithm to simultaneously transform both the state output and state duration distributions. Furthermore, we propose an HSMM-based adaptive training algorithm to simultaneously normalize the state output and state duration distributions of the average voice model. We incorporate these techniques into our HSMM-based speech synthesis system, and show their effectiveness from the results of subjective and objective evaluation tests.

This paper presents a family of rate-one quasi-orthogonal space-time block codes (QO-STBCs) for any number of transmit antennas. Full diversity of the proposed QO-STBCs is achieved via the use of constellation rotation. When the number of transmit antennas is even, these codes are delay "optimal." This property along with the quasi-orthogonality one allows the codes to have low decoding complexity. Besides, by applying lookup tables into the detection methods presented in [1] and generalizing them, two low-complexity maximum-likelihood (ML) decoders for the proposed QO-STBCs and for other existing QO-STBCs, called PMLD and QMLD, are obtained. Simulation results are provided to verify the bit error rate (BER) performances and complexities of both the proposed QO-STBCs and the proposed decoders.

In this paper, we propose a new multiuser detection scheme using Maximum Likelihood (ML) criterion and the M algorithm for Multi Carrier (MC)-Code Division Multiple Access (CDMA) systems in the down-link channel. We first describe an implementation of ML detection separating In- and Quadrature-phase components and using well-known linear filters. In the proposed algorithm, we produce hypothesis symbol vectors in a tree structure by partly changing the sub-optimum hard decisions based on the linear filter output. At each stage, we adopt the best M likely paths with respect to the true log likelihood or distance function as survivors. We determine the symbol vector which minimizes the distance function at the final stage. Although the complexity of ML detector is exponentially increasing as a function of the number of users, the proposed scheme requires by far less complexity. We demonstrate that the proposed scheme achieves equivalent Bit Error Rate (BER) performance with lower complexity in comparison with ML detector by computer simulations. Moreover we compare the proposed detection scheme with QRD-M algorithm which is based on QR decomposition combined with M algorithm.

We consider the reliability-based heuristic search methods for maximum likelihood decoding, which generate test error patterns (or, equivalently, candidate codewords) according to their heuristic values. Some studies have proposed methods for reducing the space complexity of these algorithms, which is crucially large for long block codes at medium to low signal to noise ratios of the channel. In this paper, we propose a new method for reducing the time complexity of generating candidate codewords by storing some already generated candidate codewords. Simulation results show that the increase of memory size is small.